fairchild

Markdown task backlog and project roadmap (backlog/{todo,doing,done,failed}/, backlog/ROADMAP.md) for adding, advancing, recording progress, rescuing, cancelling, retrying, failing, maintenance, or reflecting on backlog priorities and roadmap direction.

Write, review, or improve SwiftUI code. Use for SwiftUI features/refactors, state management, view composition, performance, concurrency, modern Apple APIs, or native UI quality.

Work with Claude Code configuration at global (~/.claude) or project (.claude/) level. Use when editing settings.json (permissions, hooks, statusline, model), managing MCP servers, creating agents/commands/skills, writing CLAUDE.md, setting up rules files, or configuring a new project. Determines context automatically and provides guidance on global vs project placement to avoid duplication.

Session continuity for coding work. Default to /chronicle to capture, /chronicle catchup to resume, and /chronicle pending to review open threads. Use curate, recap, wrapup, summarize, publish, insights, search, and ui only when the user explicitly needs them.

Analyze AI coding assistant usage patterns across Claude Code, Codex, and Cursor. Use when user asks about their coding usage, tool statistics, productivity patterns, skill popularity, session history, or wants to query their AI coding logs. Triggers include "usage", "how much have I used", "most used tools", "skill popularity", "coding stats", "productivity patterns".

Fix misaligned right borders in ASCII art diagrams

Customize or troubleshoot the live status line. Use when modifying statusline.sh, debugging token display, or understanding the rendering pipeline. The status line runs passively -- this skill is for working on it, not using it.

Run deterministic GUI workflows in isolated Tart macOS VMs. Provides VM lifecycle management, guest command execution via tart exec (SSH fallback), VNC interaction (clicks, keyboard, captures), batch operations, target manifests for project-specific landmarks, and clean teardown. Headless by default with optional VNC observation.

Capture deterministic screenshots of the WorkSpaces macOS app in a known UI state. Use when the user asks to "capture a release screenshot", "screenshot the app", "show the Phase 1 state", or wants evidence for a PR / release notes / design brief. Wraps the fixture-mode env vars (WORKSPACES_UI_FIXTURE, WORKSPACES_UI_FIXTURE_AGENT_STATES, and WORKSPACES_UI_FIXTURE_COMMAND_STATUSES) and the existing scripts/sidebar-capture.sh capture pipeline. Triggers on "release screenshot", "screenshot the app", "capture the app", "fixture screenshot", "ui evidence", "/release-screenshot".

Run the cofounder contributor workflow for this repo's standing personas. Use when you want April Clearwater or Plat Ironwood to sweep open PRs, discussions, and issues, then produce one structured action: proposal, discussion comment, PR review, or issue execution that pushes a branch and opens or updates a PR.

Run and maintain the canonical Workspaces performance contract across debug, installed, release, and CI environments. Use when collecting comparable before/after metrics, asserting scenario budgets, verifying installed-build parity, refreshing baselines, or classifying self-hosted perf lane failures as infrastructure versus product regressions. Prefer this skill over workspaces-optimization when the task is to operate the standardized performance system rather than investigate an unknown slowdown.

Convert an approved GitHub Discussion idea into actionable GitHub Issues and, when needed, a milestone. Use when planning endorsed ideas, turning a discussion into issue-sized work, reconciling existing planner markers, or running the Peter Planner workflow manually or from automation.

Assume one of this repo's named agent personas in the current conversation, including its repo-owned persona prompt and available memory context. Use this whenever the user types /become <persona>, says "become april", "assume Plat", "act as Peter", or asks for April Clearwater, Plat Ironwood, or Peter Planner with memory access.

Run, extend, and heal web/ app tests for this repo's Next.js dashboard. Use when the user asks to QA the web app, explore for bugs, add or heal Playwright tests, check accessibility, or measure coverage against behaviors. Triggers on "/qa", "qa", "test the web app", "explore the dashboard", "heal this test", "what's covered", "a11y check", "check the PR changes", "find regressions". Four phases — Scope (change-aware targeting), Explore (black-box heuristic testing), Author (spec-first test generation), Heal (selector-drift vs regression triage).

Execute GitHub milestones from this repo from planning through delivery. Use when the user wants to drive a milestone by number or relative target, for example `/drive m4`, `/drive to milestone 4`, `/drive next milestone`, or "deliver milestone 4". Always refresh the latest milestone, issue, PR, and workflow state from GitHub before planning, then work issues one at a time with issue updates, validation, PR management, review follow-up, and milestone retro notes.

Debug and improve WorkspaceManager performance on development hosts, CI or Tart lanes, and target machines. Use when investigating slow launch, slow typing in the Ghostty terminal, terminal focus lag, repo or workspace switching latency, or when collecting before and after evidence for a performance fix. Prefer this skill for requests like "optimize Workspaces", "why is typing slow", "performance regression", "slow machine", or "collect perf evidence".